Repurposing INCI-registered compounds as skin prebiotics for probiotic Staphylococcus epidermidis against UV-B

Repurposing existing compounds for new indications may facilitate the discovery of skin prebiotics which have not been well defined. Four compounds that have been registered by the International Nomenclature of Cosmetic Ingredients (INCI) were included to study their abilities to induce the fermentation of Staphylococcus epidermidis (S. epidermidis), a bacterial species abundant in the human skin. Liquid coco-caprylate/caprate (LCC), originally used as an emollient, effectively initiated the fermentation of S. epidermidis ATCC 12228, produced short-chain fatty acids (SCFAs), and provoked robust electricity. Application of LCC plus electrogenic S. epidermidis ATCC 12228 on mouse skin significantly reduced ultraviolet B (UV-B)-induced injuries which were evaluated by the formation of 4-hydroxynonenal (4-HNE), cyclobutane pyrimidine dimers (CPD), and skin lesions. A S. epidermidis S2 isolate with low expressions of genes encoding pyruvate dehydrogenase (pdh), and phosphate acetyltransferase (pta) was found to be poorly electrogenic. The protective action of electrogenic S. epidermidis against UV-B-induced skin injuries was considerably suppressed when mouse skin was applied with LCC in combination with a poorly electrogenic S. epidermidis S2 isolate. Exploring new indication of LCC for promoting S. epidermidis against UV-B provided an example of repurposing INCI-registered compounds as skin prebiotics.

Drug repurposing or repositioning is an effective approach to rapidly identify novel indications from known compounds 1,2 . There have been numerous successful cases of repurposed drugs, including sildenafil citrate (Viagra) as a medicine for erectile dysfunction and pulmonary arterial hypertension and raloxifene hydrochloride (Evista) as a treatment for osteoporosis in postmenopausal women 3 . Repurposing drugs, including Remdesivir 4 and Dexamethasone 5 , to discover potential forms of treatment for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is actively ongoing. Skin diseases affect half of the world's population and novel drugs for treatments of skin diseases are in high demand. The skin is the largest organ in humans and continuously protects the humans from the harmful environment 6,7 . Over 100 distinct species, which contribute to making up a total of 1 million microbes in the skin microbiome, conquer every square centimeter of human skin 6,8 . Our previous studies have demonstrated that fermentation of skin probiotic bacteria generated beneficial metabolites, such as short-chain fatty acids (SCFAs), that can attenuate skin disorders 9 . For example, Staphylococcus epidermidis (S. epidermidis), a common member in the human skin microbiome, can fermentatively metabolize carbon-rich molecules as prebiotics to yield SCFAs against pathogenic Staphylococcus aureus (S. aureus) 10 .
Prebiotics were originally defined by Gibson and Roberfroid in 1995 as nondigestible food ingredients 11 . This definition was later revised in 12 and 13 as "Glucose-based dietary fibers and non-carbohydrate substances including polyunsaturated fatty acid (PUFA) have been used as prebiotics for gut bacteria". Prebiotics for bacteria in the skin and other human organs are not yet defined. Prebiotics can provide probiotic bacteria as carbon sources to initiate the fermentation and produce SCFAs as acetate and butyrate 14 . It has been reported that oxidation of acetate or butyrate served as an electron donor to discharge electron to electron acceptors 15 . The gene-encoding proteins in the extracellular electron transfer (EET) family of homologs are believed to be present in both Gramnegative and Gram-positive bacteria 16 . Unlike Gram-negative bacteria, Gram-positive bacteria, not having an outer membrane, carry a cell envelope with a textured peptidoglycan layer and teichoic acids that are thought to be poorly electrogenic 17 . However, a flavin-based EET process has been recently identified in Gram-positive bacteria to produce electricity 16 by activation of type II NADH hydrogenase, which can catalyze electron exchange from cytosolic NADH to a quinone derivative such as quinone demethylmenaquinone (DMK) 16 . Our recent studies demonstrated that S. epidermidis can mediate glycerol as a source to generate electricity and enhanced bacterial resistance to UV-B 18 .
In this study, we selected four carbon-rich molecules that have been listed on the International Nomenclature of Cosmetic Ingredients (INCI) to examine their prebiotic activities. The liquid coco-caprylate/caprate (LCC) with C8-10 fatty acid connected to C12-C18 fatty alcohols is currently used as an emollient on ultraviolet (UV) filter absorbance 19 22 . To screen these four carbon-rich molecules for repurposing, we examined their prebiotic activities for induction of fermentation of S. epidermidis. It has been documented that S. epidermidis can mediate fermentation to down-regulate UV-B-induced inflammation in mouse skin 9 . We thus further explored the mechanism by which repurposing carbon-rich molecules as skin prebiotics influence the skin damage induced by UV-B. UV radiation which provokes free radical formation, making it a primary risk factor for skin cancer 23 . It has been reported that UV, in particular UV-B, radiation can up-regulate the local neuroendocrine axes, induce the release of hormones to circulation, activate the central hypothalamic-pituitary-adrenal axis, and reset body hemostasis against skin disorders including cancers, aging and autoimmune diseases 24 . Repurposing may facilitate the discovery of new mechanisms of action for INCI-registered ingredients as skin prebiotics against UV injuries.

Methods
Ethics statement. All animal protocols and experiments have been approved by the National Central University (NCU), Taiwan. Experiments were conducted in accordance with the protocols (NCU-106-016, 19 December 2017) of the Institutional Animal Care and Use Committee (IACUC) of National Central University (NCU). Female ICR mice (8-9 weeks old) were purchased from the National Laboratory Animal Center Taipei, Taiwan. CO 2 sedation was used to sacrifice mice in an encased chamber. All human study protocols were approved by Institutional Review Board (IRB) (No. 19-013-B1, 22 May 2019) and Ethics Committee of Landseed International Hospital, Taiwan. The methods followed for skin swab sampling procedure were carried out in accordance with relevant guidelines and regulations of IRB which was approved by Landseed International Hospital, Taiwan. Skin swabs were collected from three healthy subjects and informed consent was obtained from all study participants.
Bacterial fermentation. Staphylococcus epidermidis ATCC 12228 in tryptic soy broth (TSB) (Sigma, St. Louis, MO, USA) was cultured overnight at 37 °C. Bacterial growth was determined at 600 nm wavelength (OD 600 ). The bacterial pellet was collected after centrifugation at 5,000 × g for 10 min, resuspended with 1 × PBS and diluted to 10 7 CFU/ml before further incubation in rich media [1.5 g/l KH 2 PO 4 , 10 g/l yeast extract (Biokar Diagnostics, Beauvais, France), 2.5 g/l K 2 HPO 4 , 3 g/l TSB, and 0.002% (w/v) phenol red (Merck, Darmstadt, Germany)] at 37 °C. For fermentation, bacteria in rich media in the presence of 2% LCC, ININ, PDS or PETIS (TNJC corporation, Chiayi, Taiwan) were incubated for 12 h. The color change of phenol red from red to yellow indicated bacterial fermentation, which was quantified by measurement of OD 562 . Bacteria alone or rich media with or without LCC, ININ, PDS or PETIS served as controls. To examine the effect of LCC on the bacterial growth, S. epidermidis ATCC 12228 [10 7 colony-forming unit (CFU)/ml)] was incubated with 2% LCC or phosphate buffer saline (PBS) for 12 h at 37 °C. After incubation, bacteria were serially diluted 1: 10 0 -1:10 5 in a 96 well plate. 10 μl of serially diluted bacteria were dropped on a TSB agar plate for CFU counts.
Electricity detection. Electricity produced by S. epidermidis was detected in vitro using a chamber equipped with cathode and anode. A carbon felt (2.5 cm × 10 cm) and a carbon cloth (10 cm × 10 cm) (Homy Tech, Taoyuan, Taiwan) were utilized to fabricate anode and cathode., respectively. The cathode was wrapped up to a Nafion membrane N117 (6 cm × 6 cm) (Homy Tech), which served as a proton exchange membrane (PEM). Copper wires were used to connect anode and cathode with external resistance (1,000 Ω) 18 . S. epidermidis in the presence or absence of LCC, ININ, PDS or PETIS was pipetted on the surface of the anode. Electricity was recorded by the changes in voltage (mV) against time (min) using a digital multimeter (Lutron, DM-9962SD, Sydney, Australia). The recorded voltages in every 10 s were used for plotting a graph.
Cyclic voltammetry. The three-electrode autolab potentiostat (PGSTAT 128 N, Metrohm Autolab, Utrecht, Netherland) was used for conducting cyclic voltammetry. The screen-printed carbon electrode (SPCE) (SE-100, Zensor R&D, Taichung, Taiwan) served as a working electrode with a working area of 5 mm. The Ag/AgCl electrode and platinum electrode acted as the reference against applied potential and counter electrode, respectively. All electrodes were purchased from Metrohm Autolab. S. epidermidis (10 7 CFU/μl) in the presence or absence of 2% of LCC, ININ, PDS or PETIS was drop-coated on the surface of a working electrode. The potential windows were inspected between − 0.8 and 0.2 V at 0.005 V/s. PBS at 7.4 pH was used as an electrolyte. The potentiostat was operated using Autolab Nova 2.0 software (https ://metro hm-autol ab.com/Produ cts/Echem /Softw are/Nova. html/; Metrohm Autolab, Utrecht, Netherland).

Real-time qPCR (RT-qPCR).
RT q-PCR was used to analyze the expression of genes encoding pyruvate dehydrogenase (pdh), phosphate acetyltransferase (pta) and intracellular adhesion A (ica A) in S. epidermidis ATCC 12228 and S2 isolate. RNA (1 ng) was converted to cDNA using an iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA). The cDNA was served as a template in StepOnePlus RT PCR System (Thermo Fisher Scientific, Waltham, MA, USA), which was executed using Power SYBR Green and PCR Master Mix (Thermo Fisher Scientific). The primer-Blast tool (https ://blast .ncbi.nlm.nih.gov/Blast .cgi/; Rockville Pike, Bethesda MD, USA) from the National Center for Biotechnology Information (NCBI) was used for designing all primers. Total one step RT-PCR reaction condition was fixed for 40 cycles as follows: 95 °C for 10 min followed by 95 °C for 15 s, 60 °C for 60 s, and 72 °C for 30 s. Gene expression was normalized with the 16S rRNA gene. The cycle threshold (2 −ΔΔCt ) was implemented to analyze the relative expression of genes. The designed primers for all genes were shown in Table S2.

Gas chromatography-mass spectrometry (GC-MS) analysis. Staphylococcus epidermidis ATCC
12228 (10 7 CFU/ml) with 2% LCC in rich media was cultured for 24 h and then centrifuged at 5,000 × g for 10 min. Supernatants of bacterial culture were collected and filtered through 0.22 µm filters. The GC-MS protocol for SCFAs detection were obtained following the method 25 .
UV-B exposure. Mouse hair in the dorsal skin was removed using Nair cream (Church and Dwight, Ewing Township, NJ, USA) one day before experiments started. The dorsal skin of each mouse was exposed to 100 mJ/ cm 2 UV-B irradiation using a UV lamp (Model EB-280C, Spectronics Corp., Westbury, NY, USA) twice a week, followed by subsequent application of 10 7 CFU/ml S. epidermidis with or without 2% LCC three times per week for two weeks. The images of mouse skin were captured on day 0, 7 and 14. Skin lysates and sections from dorsal skin (1 cm 2 ) was prepared. Skin sections were stained with hematoxylin and eosin (H&E) and visualized using the Olympus BX63 microscope (Olympus, Tokyo, Japan).  www.nature.com/scientificreports/ were used to monitor bacterial electricity. Compared to media with S. epidermidis alone, media with S. epidermidis plus ININ, PDS or PETIS did not elicit high voltage changes and currents. By contrast, a robust increase in voltage changes with a peak voltage of approximately 6 mV and currents was detected in the media with S. epidermidis plus LCC (Fig. 2). The result provided evidence for repurposing INCI-registered LCC as a prebiotic to provoke the fermentation and electricity production of skin S. epidermidis bacteria.   Figure S1) and lesions (Fig. 3C), can be detected on mouse skin topically applied with LCC or S. epidermidis alone. However, the UV-B-induced the formation of 4-HNE, CPD, epidermal hyperplasia and lesions were considerably attenuated when mouse skin was topically applied with S. epidermidis plus LCC (Fig. 3, Figure S1). Topical application of PBS or LCC alone without S. epidermidis on mouse skin before UV-B irradiation exhibited the same levels of 4-HNE and CPD as well as skin lesions ( Figure S2), indicating that LCC itself was insufficient to impede the UV-B-induced skin injuries. Data above defined a novel function for the repurposed LCC in conjunction with S. epidermidis for suppressing UV-B-induced skin injuries.

INCI-registered compounds function as skin probiotics for induction of SCFA production and
The S. epidermidis S2 isolate with low expression of pdh and pta genes was poorly electrogenic. Since LCC effectively induced S. epidermidis to undergo fermentation and yield electricity, we next determined the expression of genes related to fermentation, acetate production, and biofilm formation in S. epidermidis ATCC 12228 and a S. epidermidis S2 strain isolated from human skin. The pdh gene encodes for pyruvate dehydrogenase, which catalyzes pyruvate to acetyl coenzyme A (acetyl-CoA) at the upstream site of the fermentation pathway 34 . The pta gene encoding for phosphate acetyltransferase is involved in the conversion of acetyl-CoA to acetate 35 . The icaA gene encoding for intracellular adhesion A has been well characterized in the engagement of the biofilm formation in S. epidermidis 36,37 . The 16 s rRNA sequence of S. epidermidis S2 isolate shared 99.8% identity to that of S. epidermidis ATCC 12228 (Table S1). However, the expressions of pdh and pta genes in S. epidermidis S2 isolate were much lower than those in S. epidermidis ATCC 12228 (Fig. 4A). Furthermore, the activity of LCC fermentation monitored by OD 562 reduction of phenol red-containing rich media for S. epidermidis S2 isolate was relatively low compared to that in S. epidermidis ATCC 12228 (Fig. 4B). The high expression of icaA gene (Fig. 4A) and obvious biofilms (Fig. 4D) were detected in S. epidermidis S2 isolate, but not in the non-biofilm forming bacterial strain, S. epidermidis ATCC 12228. Interestingly, unlike S. epidermidis ATCC 12228, the S. epidermidis S2 isolate that expressed low levels of pdh and pta genes was poorly electrogenic. As shown in Fig. 4C, consistent with Fig. 2, a peak voltage of approximately 6 mV was detected in media with S. epidermidis ATCC 12228 plus LCC whereas little or no voltage change was measured in S. epidermidis S2 isolate www.nature.com/scientificreports/ plus LCC. Collectively, the pdh and pta genes may participate in fermentation and electricity production of S. epidermidis triggered by LCC.

LCC plus S. epidermidis S2 isolate did not confer protection against UV-B-induced skin injuries.
To confirm the critical roles of pdh and pta genes in LCC fermentation of S. epidermidis against UV-B in vivo, dorsal skin of ICR mice was topically applied with S. epidermidis ATCC 12228 or S2 isolate in the presence of LCC before UV-B irradiation. The levels of 4-HNE and CPD in mouse skin were measured by western blot. In agreement with data in Fig. 3, after UV-B irradiation, the formation of 4-HNE, CPD and skin lesions were detected at low levels and moderate on mouse skin topically applied with S. epidermidis plus LCC. However, when mouse skin was topically applied with S. epidermidis S2 isolate plus LCC (Fig. 5A-C), UV-B induced significantly increasing levels of 4-HNE and CPD as well as skin lesions. Since the expressions of pdh and pta genes in S. epidermidis S2 isolate were much lower than those in S. epidermidis ATCC 12228 (Fig. 4), the expressions of pdh and pta genes may mediate the LCC-triggered promotion of S. epidermidis against skin injuries caused by UV-B.

Discussion
LCC is an ester obtained from the reaction of the coconut alcohol-derived fatty acids with a mixture of caprylic acid and capric acid 38 . Both caprylic acid and capric acid, as medium-chain fatty acids, can be extracted from coconut oil and have been used as ingredients for skincare formulation to protect against UV radiation 39,40 . It has been reported that addition of caprylic acid into the fermentation process enhanced acetate production 41 . Coconut oil has been used as a fuel for electricity production 42 . Although many medium-chain fatty acids exhibited potent bactericidal activities 43 , our result in Figure S3 demonstrated that LCC did not change the growth of S. epidermidis. Compared to other INCI-registered compounds (ININ, PDS and PETIS), LCC displayed higher activity in promoting fermentation and electricity production of S. epidermidis (Fig. 1). Biofilms are electroactive and can promote the electricity production of bacteria 44 . By using S. epidermidis ATCC 12228, a non-biofilm forming strain, we demonstrated that skin bacteria can produce electricity without biofilm formation on electrodes. Data in our recent publication has revealed that addition of glycerol into the S. epidermidis culture can induce fermentation and instantly produce detectable electricity, highlighting a possible mechanism that skin bacteria underwent fermentation to accumulate SCFAs as electron donors to intensify electricity 18 .
Results in Fig. 1 demonstrated that the electricity measured by changes in voltage and currents was considerably produced when the anode was pipetted with S. epidermidis ATCC 12228 plus 2% LCC. However, the electricity produced by bacteria plus LCC was largely reduced in the S. epidermidis S2 isolate which expressed www.nature.com/scientificreports/ lower levels of pdh and pta genes (Fig. 3). The data indicated that proteins corresponding to pdh and pta genes may play a role in electricity production of S. epidermidis in the presence of LCC. Electrons are derived by the reduction reaction of NAD + to NADH in the metabolic pathway of bacterial fermentation 45 . The conversion of NAD + to NADH is involved in pyruvate dehydrogenase (pdh), which initiates the process of electron transport chain 45,46 . Previous studies have shown that electricity production of Shewanella oneidensis MR-1, a representative electrochemically active bacterium (EAB) extensively studied in the laboratory, was mediated by activation of NAD + -linked PDH 47 . Phosphate acetyltransferase (pta) can catalyze the conversion of acetyl-CoA to acetate, a known electron donor. A pta knockout strain of Shewanella oneidensis strain has been used to study the effect of electricity on bacterial fermentation 48 . The protective effect of bacterial fermentation on the suppression of the UV-B-induced formation of 4-HNE and CPD was remarkably diminished when mouse skin was topically applied with LCC and S. epidermidis S2 isolate which yielded low electricity and expressed lower levels of pdh and pta genes (Fig. 5). The data suggested that SCFAs and electricity induced by LCC fermentation of S. epidermidis may synergistically provide mice protection from UV-B injuries. Future studies will include the construction of a pdh knockout S. epidermidis strain and investigation of the essential role of pdh gene in S. epidermidis for production of electricity against UV-B injuries. 4-HNE is known to be genotoxic and damages DNA by producing bulky 1,N 2 -propano-2′-deoxyguanosine adducts 49 . Here, we showed that 4-HNE and CPD were formed when mouse skin was constantly bombarded with UV-B. Results in our previous studies have demonstrated that butyrate produced by glycerol fermentation of S. epidermidis can down-regulate UV-B-induced pro-inflammatory interleukin (IL)-6 cytokines through SCFA receptor 2 (FFAR2) 50 . Thus, skin bacteria may take advantage of endogenous glycerol as a carbon source to provoke fermentation and simultaneously produced SCFAs and electrons. Besides being electron donors, SCFAs may regulate FFAR2 and/or histone deacetylases (HDAC) 51 to mitigate the inflammation induced by UV irradiation. Electrons may play a key role in neutralizing free radicals generated by UV irradiation. It has been shown that UV light can mediate nitrate, a constituent of sweat in skin, to generate free radicals 52 , which can subsequently induce lipid peroxidation to produce 4-HNE 53 . However, commensal bacteria can utilize nitrate as an electron acceptor www.nature.com/scientificreports/ for energy transduction 54 . For example, under anoxic or oxygen-depleted conduction, Pseudomonas species can mediate denitrification by using nitrate as an electron acceptor to convert nitrate to nitrogenous gases 55 . Thus, a possible mechanism behind the protective effect of S. epidermidis plus LCC on UV-B skin injuries is that LCC triggers bacteria to produce electrons, activating bacterial denitrification to reduce UV-B-induced free radicals. In summary, prebiotics for skin probiotic bacteria have been not defined by the Food and Drug Administration (FAD) even though they may become novel therapeutics for treatments of skin disorders via induction of fermentation of skin bacteria. Our study here demonstrated an approach by repurposing INCI-registered compounds as skin prebiotics and revealed their capabilities of generating electricity from skin bacteria to combat UV-B-induced skin damages. www.nature.com/scientificreports/ Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.